Bjarten/early-stopping-pytorch
A Python library providing an EarlyStopping class to monitor validation loss and halt PyTorch model training before overfitting occurs.
Not currently ranked — collecting fresh signals.
star history
The repository provides a reusable EarlyStopping class for PyTorch that tracks validation loss during training. When the loss fails to improve for a specified number of epochs (patience), training stops automatically. The class also saves model checkpoints each time validation loss decreases, enabling recovery of the best-performing model state. An MNIST example notebook demonstrates practical usage.
Frequently asked
- What is Bjarten/early-stopping-pytorch?
- A Python library providing an EarlyStopping class to monitor validation loss and halt PyTorch model training before overfitting occurs.
- Is early-stopping-pytorch open source?
- Yes — Bjarten/early-stopping-pytorch is open source, released under the MIT license.
- What language is early-stopping-pytorch written in?
- Bjarten/early-stopping-pytorch is primarily written in Jupyter Notebook.
- How popular is early-stopping-pytorch?
- Bjarten/early-stopping-pytorch has 1.3k stars on GitHub.
- Where can I find early-stopping-pytorch?
- Bjarten/early-stopping-pytorch is on GitHub at https://github.com/Bjarten/early-stopping-pytorch.